How to Minutely Transcribe a Meeting

Transcribing refers to putting together speech or thoughts into written or printed format. It is an ancient phenomenon that dates back to 3400 BCE (Before Common Era), with the Egyptians dominating the field. We have since evolved into recording and transcribing audio to written form manually.

In recent years, our advancement has leaped to recording and transcribing audio automatically. Doing this has significantly helped save time and resources, as manually transcribing one audio hour takes three to five hours.

Step-by-Step Transcription Process

Generally, three main parts are involved whenever you need to transcribe a meeting: recording the audio, creating the transcription file, and proofreading.

Recording the Audio

Many factors contribute to determining whether you end up with clear audible audio for transcription, e.g., the type of recording device, microphone, ambiance, etc. An omnidirectional microphone, for instance, would be advantageous over a tablet as it captures sounds coming from any direction far more clearly.

Another factor to consider as you transcribe meeting audio to text is the number of speakers. The more the number of participants, the more complex the transcription, especially in cases where their voices are overlapping. Using a soundproof room that allows your recording device to capture sounds coming only from the speakers is best.

The duration and type of meeting will help you determine whether you must set up cameras for video transcription or audio only. The number of recording devices required for a single session should also be predetermined, so the transcriptionist accounts for each speaker and dialogue. It's also best to ensure the recording device(s) can last the entire session without losing power or data.

Once you have considered all the above factors, conducting multiple audio tests in a similar room to where the meeting will be held is best. This helps eliminate most of the errors that can take place in the middle of your recording and ensure the desired output is acquired.

Creating the Transcription File

When required to transcribe a Zoom meeting that you are in attendance, always take notes as the recording is ongoing. Zoom has provision for recording in audio only or audio plus video. During this recording, you could use a real-time transcription tool or choose to transcribe the downloaded recording afterward.

For the real-time recording, your transcript will be ready as soon as the meeting is over. Using this method, you will be required to take notes as the session is ongoing and check on any errors made by the transcription tool.

If you choose to transcribe a downloaded recording, you can take notes during the meeting and later generate the transcript by reviewing the audio as many times as necessary. This option also frees up the transcriptionist to participate in the discussion and is ideal when you have to transcribe a Teams meeting or a physical one. 

Currently, within the Teams application, there is an option to auto-generate captions on recorded videos but only for English speakers.

Proofreading

Your transcript may be over 70% accurate, depending on the tools used, but it is rarely ever 100%. Therefore, you are required to proofread for grammatical and punctuation errors. Doing this also lets you detect mistakes caused by heavy accents or overlapping speakers. In the end, you'll achieve a near-perfect transcript.

Conclusion

Transcription has transformed the learning space, the workspace, and all other related engagements. In place of note-taking that can be tedious and sometimes highly inaccurate, when you transcribe a meeting, an accurate account of events is kept and accessible for future reference. It is highly recommended that the meeting transcription is done as soon as possible after the session to improve accuracy.